Vol. 13, No. 4 Printed in U.SA.
INFECTION AND IMMUNITY, Apr. 1976, p. 1132-1138 Copyright ©) 1976 American Society for Microbiology
Studies of Mycobacterial Antigens, with Special Reference to Mycobacterium leprae GORAN KRONVALL,I* JOHN L. STANFORD, AND GERALD P. WALSH
Armauer Hansen Research Institute, A.H.R.L, P.O. Box 1005, Addis Ababa, Ethiopia; Middlesex Hospital, London, England; and Gulf South Research Institute, New Iberia, Louisiana Received for publication 30 September 1975
Eight individual antigens were detected in soluble antigen preparations from Mycobacterium leprae bacilli by using pools of serum samples from lepromatous leprosy patients as antibody reagents in crossed immunoelectrophoresis. Two of these antigens were analyzed further. Antigen no. 1 gave an elution pattern on Sephadex G-200 corresponding to a molecular weight of 285,000. This antigen was also present in three slow-growing and eight fast-growing mycobacterial species. There was a reaction of complete identity in immunological tests using lepromatous serum pools as well as with rabbit antisera raised against M. leprae and M. smegmatis. Antigen no. 21 of M. leprae showed antigenic heterogeneity when compared with other species. Three types of antigenic determinants were detected; one, called 21A, was shared by all mycobacteria; another, called 21B, was limited to antigen no. 21 of M. leprae; a third, called 21C, was present in all mycobacteria except the leprosy bacillus. This submolecular heterogeneity may indicate a separate taxonomic position of M. leprae among the mycobacteria. Biopsies of nodules from lepromatous leprosy patients have constituted the only source of bacilli of the noncultivable bacillus Mycobacterium leprae since its discovery in 1873. In spite of inherent difficulties with this source of material such as limited amounts and the possibility of antigenic deficiency (5, 30), progress in antigenic analysis of M. leprae has been noted. Carbohydrate antigens, also found in other mycobacteria, have been identified (1, 2, 6, 7, 19, 20, 29). They have been called NEG2 (1, 2), polysaccharide I (6, 7),and the /8- and 8-antigens (19, 20). NEPR, an M. leprae-specific protein antigen with a molecular weight of more than 200,000, has been identified by Abe (1, 2). Using a plaque assay technique, a separated fraction of M. leprae containing one single protein antigen was found to be strain specific (21). Larger amounts of M. leprae bacilli grown in vivo in armadillos have recently become available (10, 28). Antigen extracts of such bacilli have been studied in two-dimensional immunodiffusion analysis (26), revealing as many as four M. leprae-specific antigens (28). In previous investigations, one M. leprae antigen was found to carry determinants unique to the leprosy bacillus (11). This antigenic heterogeneity at the molecular level indicated the possibility of an alternative approach to taxonomic studies of the leprosy bacillus. The results of such studies
reported here suggest a position of M. leprae among the mycobacteria different from the slow-growing or fast-growing mycobacterial clusters.
MATERIALS AND METHODS Mycobacterial antigen preparations. Strains of both slow-growing and fast-growing mycobacterial species (Table 1) were cultured on Sauton medium solidified with 1.5% agar. Harvested bacteria were washed three times in PBS (phosphate-buffered saline; 0.12 M NaCl, 0.03 M phosphate, pH 7.3) and suspended in 10 ml of PBS per g (wet weight). Soluble antigens were prepared by ultrasonication for 30 min, using a Branson Sonifier B-12 (Danbury, Conn.), and insoluble residues were removed by centrifugation. Protein determinations of soluble antigen preparations were performed by using FolinCiocalteau reagent (17). Antigen preparations of M. leprae were obtained from bacilli grown in vivo in armadillos (10, 28). Cutaneous nodules were extracted three times with PBS to remove soluble material by using an UltraTurrax homogenizer (Janke & Kunkel KG, IKAWerk, Staufen, Breisgau, model TP 18-10). The insoluble residue consisting of tissue material and leprosy bacilli was then dehydrated, and the bacilli were extracted by using oil-chloroform (Ricella 917chloroform, 8:2) (18, 25). Combined extracts were centrifuged, and the pellet was washed twice with acetone and then with ether. The pellet was suspended in PBS (1 ml/20 mg) and ultrasonicated. The soluble antigens were separated from insoluble resiI Present address: Institutionen for Medicinsk Mikro- dues by centrifugation. Extracts of normal armabiologi, Solvegatan 23, S-223 62 Lund, Sweden. dillo tissue were negative in crossed immunoelectro1132
TABLE
1133
ANTIGENS OF M. LEPRAE
VOL. 13, 1976
1. Immunological comparison of antigen no. 21 from different mycobacterial species using lepromatous serum pools or anti-M. smegmatis antiserum as antibody reagents Lepromatous serum poolsP
Mycobacterial strain
M. lepraeP
M. avium intracellu-
M M smegma
Anti-M. smegmatis antiserum M. lepraec
M. avium intracellu-
M. lactis
lare
lare Slow growing M. avium M. av.intrac. M. gordonae
NCTC 8551 ATCC 25153 NCTC 102 67
X X X
I I I
I I I
X X X
I I I
I I I
Fast growing M. smegmatis M. phlei M. vaccae M. diernhoferi M. ranae M. chelonei M. Iactis M. thermoresist.
NCTC 333 NCTC 8156 ATCC 15483 ATCC 19340 NCTC 2891 NCTC 946 J.S. no. 831 NCTC 10409
X X X X X X X X
I I I I I I I I
I I I I I I I I
X X X X I X X
I I I I
I I I I
Xd
X
I I
I I
X Xe I Xb Xb I M. leprae a Abbreviations: X, Cross-reactivity with partial identity; I, complete identity. b All cross-reactivities noted with lepromatous serum pools as antibody reagents were seen as a spurring by M. leprae antigen no. 21 over the others (Fig. 2C). c All cross-reactivities noted using anti-M. smegmatis antiserum were seen as a spurring by antigen no. 21 of slow and fast growers over the corresponding M. leprae antigen (Fig. 3B). d Cross-reactivities with M. chelonei were observed as a spurring over antigen no. 21 of this strain by the others.
phoresis against high-titered anti-M. leprae sera from leprosy patients. Leprosy sera. Serum samples from lepromatous patients diagnosed clinically or by histopathological examination (22-24) were pooled. LS II (lepromatous serum pool no. II) consisted of 43 individual sera and LS III of 37 serum samples from clinically diagnosed lepromatous cases. LS IV was a pool of three individual sera, two lepromatous leprosy and one borderline lepromatous case, selected for their high titers against M. smegmatis antigen no. 40 (11). LS V was made up of seven lepromatous borderline and two lepromatous sera with very high titers of antibodies against M. smegmatis antigens. Serum samples and serum pools were stored at -20 C until use. Rabbit antiserum. Antisera against M. smegmatis and M. duvalii were raised in rabbits by injection subcutaneously of 2 mg of antigen with Freund incomplete adjuvant every other week. When sufficient titers were reached, the rabbits were bled twice a week for 2 months while the immunizations were continued. Serum samples were pooled after crossed immunoelectrophoretic analysis and stored at -20 C. Crossed immunoelectrophoresis. The method of crossed immunoelectrophoresis as described by Laurell (15) was used with modifications (3, 4, 16). For comparison of antigenic cross-reactivity, the tandem technique was used. The intermediate gel technique was used to define antibody activity of serum samples in M. smegmatis, M. duvalii, or M. leprae reference systems. Antigens in eluates of chromatographic columns were detected by the fused rocket technique.
Gel filtration. Sephadex G-200 gel filtration of M.
leprae antigens was performed by using a column having an inner diameter of 25 mm and length of 950 mm. The gel was equilibrated with 0.2 M ammonium bicarbonate buffer, pH 8.0, and 3.5-ml fractions were eluted at a rate of 13.5 ml/h. The column was calibrated for determination of molecular weights. Fractions from filtration experiments with mycobacterial antigens were freeze dried, and the contents were dissolved in small volumes of immunoelectrophoresis buffer for analysis.
RESULTS Antigens detected by lepromatous serum pools. Crossed immunoelectrophoretic precipitation patterns using serum pools from lepromatous leprosy patients were obtained with antigen preparations from M. leprae, M. avium intracellulare, and M. smegmatis (Fig. 1). The latter two were taken as representatives of slow-growing and fast-growing mycobacteria, respectively. By including increasing concentrations of lepromatous serum pools in the intermediate gel of an M. smegmatis reference system described earlier (11), four precipitin lines could be assigned corresponding numbers: antigens no. 1, 21, 40, and 41 (Fig. 1C). Using the tandem crossed immunoelectrophoretic technique, the same antigens were also identified in M. avium intracellulare and M. leprae antigen preparations (Fig. 1A, B). The other precipitin lines of M. leprae were designated
1134
INFECT. IMMUN.
KRONVALL, STANFORD, AND WALSH
letters d, e, f, and g (Fig. 1A). Further comparibetween M. leprae, M. avium intracellulare, and M. smegmatis indicated that antigens d, e, and f were present in M. avium intracellulare and that antigens d and e were present in M. smegmatis (Fig. 1). Comparative studies of antigen no. 1 in mycobacterial species. Lepromatous S III was used as an antibody source at a concentration of 2% in the antibody gel for comparisons between sons
LS containing agarose
AX
+ LS containing agarose 2nd dim.
+ LS containing agarose 2nd dim.
+ 2nd dim
f Osi40 i +l 1d
~~~lstdim.+
0
antigen no. 1 of different mycobacterial species. There was a reaction of complete identity between antigen no. 1 of M. leprae and that of M. avium, M. avium intracellulare, M. gordonae, M. smegmatis, M. vaccae, M. diernhoferi, M. duvalii, M. rhodesiae, and M. chitae (Fig. 2A). Similar experiments with M. avium intracellulare in tandem crossed immunoelectrophoresis with M. gordonae, M. phlei, and M. ranae as well as M. smegmatis in tandem with M. av-
0lst dimn.+
0lst dim +
A. B. C. FIG. 1. Crossed immunoelectrophoresis of mycobacterial antigen preparations against pools of serum samples from lepromatous leprosy patients. (A) M. leprae; (B) M. avium intracellulare; (C) M. smegmatis antigens. Identification of antigens as indicated by numbers and letters was performed in separate experiments using the tandem crossed immunoelectrophoretic technique.
Wm&, 4.
.VI/
LS
I
+
LS
+
LS
+
EIA
V
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I
III
|M.I_00M.d.
_
PS Ja.
+ lo
I
M. S
+
+ Raj |.IoW1JS
A. B. 1 and no. 21 in FIG. 2. Examples of tandem crossed immunoelectrophoretic analysis of antigen different mycobacterial strains, using lepromatous serum pools (LS) as antibody reagents. Relevant antigens are shown in drawings with actual photographs above. (A) Identity reaction between antigen no. 1 of M. leprae and M. duvalii; (B) identity reaction between antigen no. 21 of M. avium intracellulare and M. smegmatis; (C) reaction ofpartial identity between antigen no. 21 of M. leprae and M. avium intracellulare. Cn
no.
VOL. 13, 1976
ium intracellulare, M. diernhoferi, and M. duvalii showed complete identity between all antigens no. 1. An antigen preparation from M. kansasii indicated slight antigenic deficiency of antigen no. 1 when compared with both M. avium intracellulare and M. smegmatis as judged by a faint spur formation by antigen no. 1 of these strains over M. kansasii. Identity reactions obtained with lepromatous serum pools were confirmed in comparisons between M. leprae, M. avium intracellulare, and M. smegmatis using rabbit antisera raised against M. leprae antigens and against M. duvalii antigens. Sephadex G-200 gel filtration experiments indicated a molecular weight of antigen no. 1 from M. leprae of about 285,000. Antigen no. 21 in cultivable mycobacteria. Antigen no. 21 of M. avium intracellulare was compared with the corresponding antigen of both slow-growing and fast-growing mycobacterial species using 5 to 20% LS II, III, and IV as antibody reagents. The antigen was detected in three slow-growing and eight fast-growing mycobacterial species (Table 1). In all instances there was a reaction of complete immunological identity (Fig. 2B, Table 1). No antigen no. 21 was detected in direct tandem precipitation tests with M. kansasii or M. duvalii antigen preparations. Using an inhibition technique with increasing amounts included in the intermediate gel, the presence of small amounts of antigen no. 21 could be detected in M. duvalii antigen extracts. Antigen no. 21 of M. smegmatis was also compared with that of other strains. In all cases there was a reaction of complete identity between the antigens (Table 1). M. leprae-specific determinants on antigen no. 21. When antigens no. 21 of M. avium intracellulare and M. smegmatis were compared with that of M. leprae in tandem crossed
electrophoresis using lepromatous serum pools, a reaction of partial immunological identity was observed. Antigen no. 21 of M. leprae spurred over those of the other two strains (Fig. 2C). Prominent spurring was seen using all four lepromatous serum pools. The partial identity was confirmed in inhibition experiments. M. smegmatis antigen was included in excess in the intermediate gel in tandem crossed immunoelectrophoresis with M. leprae and M. avium intracellulare. The precipitation line of antigen no. 21 of the latter strain was completely inhibited, whereas the M. leprae no. 21 line was essentially intact. Comparisons of M. leprae antigen no. 21 were also made with those of other mycobacterial strains (Table 1). In all these experiments there
ANTIGENS OF M. LEPRAE
1135
was a marked spurring by antigen no. 21 from the leprosy bacillus over the other ones. The results of these immunological studies of antigen no. 21 indicate the existence of at least two types of antigenic determinants. One type is shared by all mycobacterial species, and the other one is detected only in M. keprae. Determinants on antigen no. 21 common to mycobacteria except M. leprae. Rabbit antiserum raised against M. smegmatis antigens has previously been used in a reference system showing more than 50 precipitation lines against M. smegmatis, 39 of which were numbered, including antigen no. 21 (11). The same antiserum gave precipitin reactions also against antigen no. 21 of other mycobacterial species. Using anti-M. smegmatis antiserum as an antibody reagent in tandem crossed immunoelectrophoretic comparisons, this antigen of M. avium intracellulare as well as of M. lactis was shown to give reactions of complete identity with three slow-growing and six fast-growing mycobacterial species (Fig. 3A, Table 1). Identity reactions were obtained in tests between M. gordonae and three fast growers (M. phlei, M. diernhoferi, and M. lactis) as well as between M. avium and three fast growers (M. smegmatis, M. phlei, and M. vaccae). M. chelonei gave discordant results. In view of the regular pattern against lepromatous serum pools as antibody reagents (Table 1), the results obtained against anti-M. smegmatis antiserum would not imply any closer link with M. leprae. The phenomenon is being studied further. Antigen no. 21 of M. leprae was compared in similar experiments using anti-M. smegmatis antiserum with the corresponding antigens of cultivable slow-growing and fast-growing mycobacteria (Fig. 3B, Table 1). A reaction of partial immunological identity was noted in all cases, with a spurring by other strains over the M. leprae antigen. These results indicate the existence of antigenic determinants that are present in all mycobacterial strains tested with the exception of the leprosy bacillus. The findings suggest a closer relationship between the cultivable mycobacteria than between any of these and M. leprae. Antigen no. 40 in mycobacterial strains. One of the lepromatous serum pools, LS IV, consisted of three selected individual serum samples with precipitating antibodies against antigen no. 40 (11). When this lepromatous pool was used as a reagent in comparative immunological studies, a reaction of complete identity was noted between antigen no. 40 from M. leprae, M. avium intracellulare, and M. smegmatis. The occurrence of this antigen in all three
1136
INFECT. IMMUN.
KRONVALL, STANFORD, AND WALSH
-.1
-1
.1
:..-
+
a-M.s.
a-M.s.
+
I~~~~ -~~~~~~ /
~~~
00 .ai.M.v.
+ 00
A.
M.I. M.a.i.
->-'
+
B.
FIG. 3. Examples of tandem crossed immunoelectrophoretic analysis of antigen no. 21 in mycobacteria, using a rabbit anti-M. smegmatis antiserum (a-M.s.). (A) Complete identity between antigen no. 21 of M. avium intracellulare and M. vaccae; (B) reaction ofpartial identity between antigen no. 21 of M. leprae and M. avium intracellulare.
species (Table 2). Using antiserum raised against a fast-growing mycobacterium, M. smegmatis, a third type of antigenic determinant was detected on no. 21 molecules. These determinants, called 21C, were detected in DISCUSSION slow-growing as well as fast-growing species Antigenic heterogeneity of a mycobacterial but were absent in M. leprae (Table 2). Occaantigen was detected in previous studies (11). sional crossed immunoelectrophoretic runs usThis antigen was labeled no. 21 in a reference ing this antiserum showed a very faint spur system for M. smegmatis, and the number was formation by no. 21 molecules from fast growers also used for the corresponding antigens of over slow growers, but the method was not other mycobacterial species. Antigen no. 21 of sensitive enough to document this suggested M. leprae was frequently involved in antibody difference. responses in lepromatous leprosy patients (11). The antigenic heterogeneity of no. 21 moleIn the present investigations the existence of cules from mycobacteria indicates a unique pounique antigenic determinants on no. 21 mole- sition of the M. leprae antigen as compared cules from M. leprae was confirmed. Such de- with the other ones studied. Determinants laterminants, called 21B, were absent in both beled 21C are group specific, suggesting that slow-growing and fast-growing mycobacterial 21B determinants more or less also express strains with complete identity suggests that antigen no. 40, like antigen no. 1, is shared by all mycobacteria (26).
VOL. 13, 1976
ANTIGENS OF M. LEPRAE
TABLE 2. Occurrence of three different types of determinants on antigen no. 21 in various mycobacterial strains
1137
ACKNOWLEDGMENTS We wish to thank Monica L6fgren, Arvid Nygaard and Kathy Joy for their excellent technical assistance.
Determinant Organism 21A
21B
21C
+
-
+
Fast-growing mycobacteria (8 species)
+
-
+
M. leprae
+
+
Slow growing mycobacteria (3 species)
specificity (Table 2). Immunological studies of the heterogeneity of individual mammalian antigens have been found to reflect the taxonomic positions of different species. Our results may have an analogous interpretation regarding the taxonomic position of M. leprae. The leprosy bacillus may not have any closer relationship with the representatives of the slow-growing and fast-growing mycobacteria studied. Its position in the genus Mycobacterium may therefore be separate from these other mycobacterial species. Similar results have been obtained recently by using a more conventional approach (26, 27) and are in line with earlier cytochemical investigations (9). A closer link with mycobacteria than with related genera is still suggested by studies of mycolic acids (8) as well as by earlier investigations (12-14). M. lepraemurium is the other known noncultivable mycobacterium causing disease. Studies of this bacillus have shown a close relationship with slow-growing mycobacteria, particularly with M. avium (25). Our results would suggest that the human leprosy bacillus is no more related to the mouse leprosy bacillus than to any of the slow-growing cultivable mycobacteria, including M. tuberculosis. The taxonomic validity and homogeneity of the so-called slow-growing as well as fast-growing mycobacterial groups would permit such a conclusion (26). Identification of M. leprae-specific antigens have been previously reported. Abe described a high-molecular-weight protein (NEPR) unique to the leprosy bacillus (1, 2). Stanford et al. have detected four different antigens that are found only in M. leprae (27). Since the defect in cellular immune resistance in lepromatous leprosy involves only the leprosy bacillus, it follows that the antigen(s) responsible for immunity must be M. leprae specific. Such antigens therefore deserve special attention regarding their role in humoral as well as cellular immunity in the various forms of leprosy. group
LITERATURE CITED 1. Abe, M. 1970. Studies on the antigenic specificity of Mycobacterium leprae. Demonstration of soluble antigens in leprosy modules by immunodiffusion. Int. J. Lepr. 38:113-125. 2. Abe, M., F. Minagawa, Y. Yoshino, and K. Okamura. 1972. Studies on the antigenic specificity of Mycobacterium leprae. Purification and immunological characterization of the soluble antigen in leprosy nodules. Int. J. Lepr. 40:107-117. 3. Axelsen, N. H. 1973. Quantitative immunoelectrophoretic methods as tools for a polyvalent approach to standardization in the immunochemistry of Candida albicans. Infect. Immun. 7:949-960. 4. Axelsen, N. H., J. Krall, and B. Weeke. 1973. Manual of quantitative immunoelectrophoresis. Methods and applications. Scand. J. Immunol. 2 (Suppl.):1-169. 5. Bekierkunst, A. 1968. Tubercle bacilli grown in vivo. Ann. N.Y. Acad. Sci. 154:79-87. 6. Estrada-Parra, S. 1972. Immunochemical identification of a defined antigen of Mycobacterium leprae. Infect. Immun. 5:258-259. 7. Estrada-Parra, S., S. Calder6n-Manes, M. SalazarMallen, and M.-E. Amezcua. 1966. Isolation of a group-specific polysaccharide from tissues infected with Mycobacterium leprae. Int. J. Lepr. 34:294-297. 8. Etemadi, A. H., and J. Convit. 1974. Mycolic acids from "noncultivable" mycobacteria. Infect. Immun. 10:236-239. 9. Fisher, C. A., and L. Barksdale. 1973. Cytochemical reactions of human leprosy bacilli and mycobacteria: ultrastructural implications. J. Bacteriol. 113:13891399. 10. Kirchheimer, W. F., and E. E. Storrs. 1971. Attempts to establish the armadillo (Dasypus novemcinctus Linn.) as a model for the study of leprosy. I. Report of lepromatoid leprosy in an experimentally infected armadillo. Int. J. Lepr. 39:693-702. 11. Kronvall, G., G. Bjune, J. Stanford, S. Menzel, and D. Samuel. 1975. Mycobacterial antigens in antibody responses of leprosy patients. Int. J. Lepr., in press. 12. Kwapinski, J. B. G., A. Alcasid, E. H. Kwapinski, and V. Nairn. 1972. The immunology of cytoplasmic antigens of mycobacteria. Can. J. Microbiol. 18:12011211. 13. Kwapinski, J. B. G., J. 0. de Almeida, and E. H. Kwapinski. 1972. Immunological analysis of Mycobacterium leprae using cross reactions with cytoplasmic antigens of the Actinomycetales. Can. J. Microbiol. 18:391-396. 14. Kwapinski, J. B. G., J. 0. de Almeida, and E. H. Kwapinski. 1972. Immunological determination of Mycobacterium leprae by means of cytoplasmic antigens. Bull. W.H.O. 46:509-513. 15. Laurell, C.-B. 1965. Antigen-antibody crossed electrophoresis. Anal. Biochem. 10:358-361. 16. Laurell, C.-B. 1972. Electrophoretic and electroimmunochemical analysis of proteins. Scand. J. Clin. Lab. Invest. 29:(Suppl):1-136. 17. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 18. Mudd, S. 1925. A study by new methods of the surfaces of normal and sensitized acid-fast bacteria. Proc. Soc. Exp. Biol. Med. 23:569-572. 19. Navalkar, R. G. 1971. Immunologic analysis of Mycobacterium leprae antigens by means of diffusion-in-
1138
KRONVALL, STANFORD, AND WALSH
gel methods. Int. J. Lepr. 39:105-112. 20. Navalkar, R. G. 1973. Immunologic studies on leprosy. 2. Antigenic studies of Mycobacterium leprae. Z. Tropenmed. Parasitol. 24:66-72. 21. Navalkar, R. G., P. J. Patel, R. R. Dalvi, and L. Levy. 1974. Immune response to Mycobacterium leprae: plaque-forming cells in mice. Infect. Immun. 10:13021306. 22. Ridley, D. S., and W. H. Jopling. 1962. A classification of leprosy for research purposes. Lepr. Rev. 33:119128. 23. Ridley, D. S., and W. H. Jopling. 1966. Classification of leprosy according to immunity. A five-group system. Int. J. Lepr. 34:255-273. 24. Ridley, D. S., and M. F. R. Waters. 1969. Significance of variations within the lepromatous group. Lepr. Rev. 40:143-152. 25. Stanford, J. L. 1973. An immunodiffusion analysis of
INFECT. IMMUN. 26.
27.
28. 29. 30.
Mycobacterium lepraemurium Marchoux and Sorel. J. Med. Microbiol. 6:435-439. Stanford, J. L. 1973. Immunodiffusion analysis-a rational basis for the taxonomy of mycobacteria. Ann. Soc. Belge Med. Trop. 53:321-330. Stanford, J. L., G. A. W. Rook, J. Convit, T. Godal, G. Kronvall, R. J. W. Rees, and G. P. Walsh. Preliminary taxonomic studies on the leprosy bacillus. Br. J. Exp. Pathol. 56:579-585. Storrs, E. E., G. P. Walsh, H. P. Burchfield, and C. H. Binford. 1974. Leprosy in the armadillos, new model for biomedical research. Science 183:851-852. Tuma, M., and C. Silva. 1962. Antigenic relationship between the Hansen bacillus and other mycobacteria. Int. J. Lepr. 30:71-76. Turcotte, R. 1973. Antigenic comparison between in vivo and in vitro grown tubercle bacilli. Can. J. Microbiol. 19:925-930.
INFECTION AND IMMUNITY, Apr. 1976, p. 1132-1138 Copyright ©) 1976 American Society for Microbiology
Studies of Mycobacterial Antigens, with Special Reference to Mycobacterium leprae GORAN KRONVALL,I* JOHN L. STANFORD, AND GERALD P. WALSH
Armauer Hansen Research Institute, A.H.R.L, P.O. Box 1005, Addis Ababa, Ethiopia; Middlesex Hospital, London, England; and Gulf South Research Institute, New Iberia, Louisiana Received for publication 30 September 1975
Eight individual antigens were detected in soluble antigen preparations from Mycobacterium leprae bacilli by using pools of serum samples from lepromatous leprosy patients as antibody reagents in crossed immunoelectrophoresis. Two of these antigens were analyzed further. Antigen no. 1 gave an elution pattern on Sephadex G-200 corresponding to a molecular weight of 285,000. This antigen was also present in three slow-growing and eight fast-growing mycobacterial species. There was a reaction of complete identity in immunological tests using lepromatous serum pools as well as with rabbit antisera raised against M. leprae and M. smegmatis. Antigen no. 21 of M. leprae showed antigenic heterogeneity when compared with other species. Three types of antigenic determinants were detected; one, called 21A, was shared by all mycobacteria; another, called 21B, was limited to antigen no. 21 of M. leprae; a third, called 21C, was present in all mycobacteria except the leprosy bacillus. This submolecular heterogeneity may indicate a separate taxonomic position of M. leprae among the mycobacteria. Biopsies of nodules from lepromatous leprosy patients have constituted the only source of bacilli of the noncultivable bacillus Mycobacterium leprae since its discovery in 1873. In spite of inherent difficulties with this source of material such as limited amounts and the possibility of antigenic deficiency (5, 30), progress in antigenic analysis of M. leprae has been noted. Carbohydrate antigens, also found in other mycobacteria, have been identified (1, 2, 6, 7, 19, 20, 29). They have been called NEG2 (1, 2), polysaccharide I (6, 7),and the /8- and 8-antigens (19, 20). NEPR, an M. leprae-specific protein antigen with a molecular weight of more than 200,000, has been identified by Abe (1, 2). Using a plaque assay technique, a separated fraction of M. leprae containing one single protein antigen was found to be strain specific (21). Larger amounts of M. leprae bacilli grown in vivo in armadillos have recently become available (10, 28). Antigen extracts of such bacilli have been studied in two-dimensional immunodiffusion analysis (26), revealing as many as four M. leprae-specific antigens (28). In previous investigations, one M. leprae antigen was found to carry determinants unique to the leprosy bacillus (11). This antigenic heterogeneity at the molecular level indicated the possibility of an alternative approach to taxonomic studies of the leprosy bacillus. The results of such studies
reported here suggest a position of M. leprae among the mycobacteria different from the slow-growing or fast-growing mycobacterial clusters.
MATERIALS AND METHODS Mycobacterial antigen preparations. Strains of both slow-growing and fast-growing mycobacterial species (Table 1) were cultured on Sauton medium solidified with 1.5% agar. Harvested bacteria were washed three times in PBS (phosphate-buffered saline; 0.12 M NaCl, 0.03 M phosphate, pH 7.3) and suspended in 10 ml of PBS per g (wet weight). Soluble antigens were prepared by ultrasonication for 30 min, using a Branson Sonifier B-12 (Danbury, Conn.), and insoluble residues were removed by centrifugation. Protein determinations of soluble antigen preparations were performed by using FolinCiocalteau reagent (17). Antigen preparations of M. leprae were obtained from bacilli grown in vivo in armadillos (10, 28). Cutaneous nodules were extracted three times with PBS to remove soluble material by using an UltraTurrax homogenizer (Janke & Kunkel KG, IKAWerk, Staufen, Breisgau, model TP 18-10). The insoluble residue consisting of tissue material and leprosy bacilli was then dehydrated, and the bacilli were extracted by using oil-chloroform (Ricella 917chloroform, 8:2) (18, 25). Combined extracts were centrifuged, and the pellet was washed twice with acetone and then with ether. The pellet was suspended in PBS (1 ml/20 mg) and ultrasonicated. The soluble antigens were separated from insoluble resiI Present address: Institutionen for Medicinsk Mikro- dues by centrifugation. Extracts of normal armabiologi, Solvegatan 23, S-223 62 Lund, Sweden. dillo tissue were negative in crossed immunoelectro1132
TABLE
1133
ANTIGENS OF M. LEPRAE
VOL. 13, 1976
1. Immunological comparison of antigen no. 21 from different mycobacterial species using lepromatous serum pools or anti-M. smegmatis antiserum as antibody reagents Lepromatous serum poolsP
Mycobacterial strain
M. lepraeP
M. avium intracellu-
M M smegma
Anti-M. smegmatis antiserum M. lepraec
M. avium intracellu-
M. lactis
lare
lare Slow growing M. avium M. av.intrac. M. gordonae
NCTC 8551 ATCC 25153 NCTC 102 67
X X X
I I I
I I I
X X X
I I I
I I I
Fast growing M. smegmatis M. phlei M. vaccae M. diernhoferi M. ranae M. chelonei M. Iactis M. thermoresist.
NCTC 333 NCTC 8156 ATCC 15483 ATCC 19340 NCTC 2891 NCTC 946 J.S. no. 831 NCTC 10409
X X X X X X X X
I I I I I I I I
I I I I I I I I
X X X X I X X
I I I I
I I I I
Xd
X
I I
I I
X Xe I Xb Xb I M. leprae a Abbreviations: X, Cross-reactivity with partial identity; I, complete identity. b All cross-reactivities noted with lepromatous serum pools as antibody reagents were seen as a spurring by M. leprae antigen no. 21 over the others (Fig. 2C). c All cross-reactivities noted using anti-M. smegmatis antiserum were seen as a spurring by antigen no. 21 of slow and fast growers over the corresponding M. leprae antigen (Fig. 3B). d Cross-reactivities with M. chelonei were observed as a spurring over antigen no. 21 of this strain by the others.
phoresis against high-titered anti-M. leprae sera from leprosy patients. Leprosy sera. Serum samples from lepromatous patients diagnosed clinically or by histopathological examination (22-24) were pooled. LS II (lepromatous serum pool no. II) consisted of 43 individual sera and LS III of 37 serum samples from clinically diagnosed lepromatous cases. LS IV was a pool of three individual sera, two lepromatous leprosy and one borderline lepromatous case, selected for their high titers against M. smegmatis antigen no. 40 (11). LS V was made up of seven lepromatous borderline and two lepromatous sera with very high titers of antibodies against M. smegmatis antigens. Serum samples and serum pools were stored at -20 C until use. Rabbit antiserum. Antisera against M. smegmatis and M. duvalii were raised in rabbits by injection subcutaneously of 2 mg of antigen with Freund incomplete adjuvant every other week. When sufficient titers were reached, the rabbits were bled twice a week for 2 months while the immunizations were continued. Serum samples were pooled after crossed immunoelectrophoretic analysis and stored at -20 C. Crossed immunoelectrophoresis. The method of crossed immunoelectrophoresis as described by Laurell (15) was used with modifications (3, 4, 16). For comparison of antigenic cross-reactivity, the tandem technique was used. The intermediate gel technique was used to define antibody activity of serum samples in M. smegmatis, M. duvalii, or M. leprae reference systems. Antigens in eluates of chromatographic columns were detected by the fused rocket technique.
Gel filtration. Sephadex G-200 gel filtration of M.
leprae antigens was performed by using a column having an inner diameter of 25 mm and length of 950 mm. The gel was equilibrated with 0.2 M ammonium bicarbonate buffer, pH 8.0, and 3.5-ml fractions were eluted at a rate of 13.5 ml/h. The column was calibrated for determination of molecular weights. Fractions from filtration experiments with mycobacterial antigens were freeze dried, and the contents were dissolved in small volumes of immunoelectrophoresis buffer for analysis.
RESULTS Antigens detected by lepromatous serum pools. Crossed immunoelectrophoretic precipitation patterns using serum pools from lepromatous leprosy patients were obtained with antigen preparations from M. leprae, M. avium intracellulare, and M. smegmatis (Fig. 1). The latter two were taken as representatives of slow-growing and fast-growing mycobacteria, respectively. By including increasing concentrations of lepromatous serum pools in the intermediate gel of an M. smegmatis reference system described earlier (11), four precipitin lines could be assigned corresponding numbers: antigens no. 1, 21, 40, and 41 (Fig. 1C). Using the tandem crossed immunoelectrophoretic technique, the same antigens were also identified in M. avium intracellulare and M. leprae antigen preparations (Fig. 1A, B). The other precipitin lines of M. leprae were designated
1134
INFECT. IMMUN.
KRONVALL, STANFORD, AND WALSH
letters d, e, f, and g (Fig. 1A). Further comparibetween M. leprae, M. avium intracellulare, and M. smegmatis indicated that antigens d, e, and f were present in M. avium intracellulare and that antigens d and e were present in M. smegmatis (Fig. 1). Comparative studies of antigen no. 1 in mycobacterial species. Lepromatous S III was used as an antibody source at a concentration of 2% in the antibody gel for comparisons between sons
LS containing agarose
AX
+ LS containing agarose 2nd dim.
+ LS containing agarose 2nd dim.
+ 2nd dim
f Osi40 i +l 1d
~~~lstdim.+
0
antigen no. 1 of different mycobacterial species. There was a reaction of complete identity between antigen no. 1 of M. leprae and that of M. avium, M. avium intracellulare, M. gordonae, M. smegmatis, M. vaccae, M. diernhoferi, M. duvalii, M. rhodesiae, and M. chitae (Fig. 2A). Similar experiments with M. avium intracellulare in tandem crossed immunoelectrophoresis with M. gordonae, M. phlei, and M. ranae as well as M. smegmatis in tandem with M. av-
0lst dimn.+
0lst dim +
A. B. C. FIG. 1. Crossed immunoelectrophoresis of mycobacterial antigen preparations against pools of serum samples from lepromatous leprosy patients. (A) M. leprae; (B) M. avium intracellulare; (C) M. smegmatis antigens. Identification of antigens as indicated by numbers and letters was performed in separate experiments using the tandem crossed immunoelectrophoretic technique.
Wm&, 4.
.VI/
LS
I
+
LS
+
LS
+
EIA
V
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I
III
|M.I_00M.d.
_
PS Ja.
+ lo
I
M. S
+
+ Raj |.IoW1JS
A. B. 1 and no. 21 in FIG. 2. Examples of tandem crossed immunoelectrophoretic analysis of antigen different mycobacterial strains, using lepromatous serum pools (LS) as antibody reagents. Relevant antigens are shown in drawings with actual photographs above. (A) Identity reaction between antigen no. 1 of M. leprae and M. duvalii; (B) identity reaction between antigen no. 21 of M. avium intracellulare and M. smegmatis; (C) reaction ofpartial identity between antigen no. 21 of M. leprae and M. avium intracellulare. Cn
no.
VOL. 13, 1976
ium intracellulare, M. diernhoferi, and M. duvalii showed complete identity between all antigens no. 1. An antigen preparation from M. kansasii indicated slight antigenic deficiency of antigen no. 1 when compared with both M. avium intracellulare and M. smegmatis as judged by a faint spur formation by antigen no. 1 of these strains over M. kansasii. Identity reactions obtained with lepromatous serum pools were confirmed in comparisons between M. leprae, M. avium intracellulare, and M. smegmatis using rabbit antisera raised against M. leprae antigens and against M. duvalii antigens. Sephadex G-200 gel filtration experiments indicated a molecular weight of antigen no. 1 from M. leprae of about 285,000. Antigen no. 21 in cultivable mycobacteria. Antigen no. 21 of M. avium intracellulare was compared with the corresponding antigen of both slow-growing and fast-growing mycobacterial species using 5 to 20% LS II, III, and IV as antibody reagents. The antigen was detected in three slow-growing and eight fast-growing mycobacterial species (Table 1). In all instances there was a reaction of complete immunological identity (Fig. 2B, Table 1). No antigen no. 21 was detected in direct tandem precipitation tests with M. kansasii or M. duvalii antigen preparations. Using an inhibition technique with increasing amounts included in the intermediate gel, the presence of small amounts of antigen no. 21 could be detected in M. duvalii antigen extracts. Antigen no. 21 of M. smegmatis was also compared with that of other strains. In all cases there was a reaction of complete identity between the antigens (Table 1). M. leprae-specific determinants on antigen no. 21. When antigens no. 21 of M. avium intracellulare and M. smegmatis were compared with that of M. leprae in tandem crossed
electrophoresis using lepromatous serum pools, a reaction of partial immunological identity was observed. Antigen no. 21 of M. leprae spurred over those of the other two strains (Fig. 2C). Prominent spurring was seen using all four lepromatous serum pools. The partial identity was confirmed in inhibition experiments. M. smegmatis antigen was included in excess in the intermediate gel in tandem crossed immunoelectrophoresis with M. leprae and M. avium intracellulare. The precipitation line of antigen no. 21 of the latter strain was completely inhibited, whereas the M. leprae no. 21 line was essentially intact. Comparisons of M. leprae antigen no. 21 were also made with those of other mycobacterial strains (Table 1). In all these experiments there
ANTIGENS OF M. LEPRAE
1135
was a marked spurring by antigen no. 21 from the leprosy bacillus over the other ones. The results of these immunological studies of antigen no. 21 indicate the existence of at least two types of antigenic determinants. One type is shared by all mycobacterial species, and the other one is detected only in M. keprae. Determinants on antigen no. 21 common to mycobacteria except M. leprae. Rabbit antiserum raised against M. smegmatis antigens has previously been used in a reference system showing more than 50 precipitation lines against M. smegmatis, 39 of which were numbered, including antigen no. 21 (11). The same antiserum gave precipitin reactions also against antigen no. 21 of other mycobacterial species. Using anti-M. smegmatis antiserum as an antibody reagent in tandem crossed immunoelectrophoretic comparisons, this antigen of M. avium intracellulare as well as of M. lactis was shown to give reactions of complete identity with three slow-growing and six fast-growing mycobacterial species (Fig. 3A, Table 1). Identity reactions were obtained in tests between M. gordonae and three fast growers (M. phlei, M. diernhoferi, and M. lactis) as well as between M. avium and three fast growers (M. smegmatis, M. phlei, and M. vaccae). M. chelonei gave discordant results. In view of the regular pattern against lepromatous serum pools as antibody reagents (Table 1), the results obtained against anti-M. smegmatis antiserum would not imply any closer link with M. leprae. The phenomenon is being studied further. Antigen no. 21 of M. leprae was compared in similar experiments using anti-M. smegmatis antiserum with the corresponding antigens of cultivable slow-growing and fast-growing mycobacteria (Fig. 3B, Table 1). A reaction of partial immunological identity was noted in all cases, with a spurring by other strains over the M. leprae antigen. These results indicate the existence of antigenic determinants that are present in all mycobacterial strains tested with the exception of the leprosy bacillus. The findings suggest a closer relationship between the cultivable mycobacteria than between any of these and M. leprae. Antigen no. 40 in mycobacterial strains. One of the lepromatous serum pools, LS IV, consisted of three selected individual serum samples with precipitating antibodies against antigen no. 40 (11). When this lepromatous pool was used as a reagent in comparative immunological studies, a reaction of complete identity was noted between antigen no. 40 from M. leprae, M. avium intracellulare, and M. smegmatis. The occurrence of this antigen in all three
1136
INFECT. IMMUN.
KRONVALL, STANFORD, AND WALSH
-.1
-1
.1
:..-
+
a-M.s.
a-M.s.
+
I~~~~ -~~~~~~ /
~~~
00 .ai.M.v.
+ 00
A.
M.I. M.a.i.
->-'
+
B.
FIG. 3. Examples of tandem crossed immunoelectrophoretic analysis of antigen no. 21 in mycobacteria, using a rabbit anti-M. smegmatis antiserum (a-M.s.). (A) Complete identity between antigen no. 21 of M. avium intracellulare and M. vaccae; (B) reaction ofpartial identity between antigen no. 21 of M. leprae and M. avium intracellulare.
species (Table 2). Using antiserum raised against a fast-growing mycobacterium, M. smegmatis, a third type of antigenic determinant was detected on no. 21 molecules. These determinants, called 21C, were detected in DISCUSSION slow-growing as well as fast-growing species Antigenic heterogeneity of a mycobacterial but were absent in M. leprae (Table 2). Occaantigen was detected in previous studies (11). sional crossed immunoelectrophoretic runs usThis antigen was labeled no. 21 in a reference ing this antiserum showed a very faint spur system for M. smegmatis, and the number was formation by no. 21 molecules from fast growers also used for the corresponding antigens of over slow growers, but the method was not other mycobacterial species. Antigen no. 21 of sensitive enough to document this suggested M. leprae was frequently involved in antibody difference. responses in lepromatous leprosy patients (11). The antigenic heterogeneity of no. 21 moleIn the present investigations the existence of cules from mycobacteria indicates a unique pounique antigenic determinants on no. 21 mole- sition of the M. leprae antigen as compared cules from M. leprae was confirmed. Such de- with the other ones studied. Determinants laterminants, called 21B, were absent in both beled 21C are group specific, suggesting that slow-growing and fast-growing mycobacterial 21B determinants more or less also express strains with complete identity suggests that antigen no. 40, like antigen no. 1, is shared by all mycobacteria (26).
VOL. 13, 1976
ANTIGENS OF M. LEPRAE
TABLE 2. Occurrence of three different types of determinants on antigen no. 21 in various mycobacterial strains
1137
ACKNOWLEDGMENTS We wish to thank Monica L6fgren, Arvid Nygaard and Kathy Joy for their excellent technical assistance.
Determinant Organism 21A
21B
21C
+
-
+
Fast-growing mycobacteria (8 species)
+
-
+
M. leprae
+
+
Slow growing mycobacteria (3 species)
specificity (Table 2). Immunological studies of the heterogeneity of individual mammalian antigens have been found to reflect the taxonomic positions of different species. Our results may have an analogous interpretation regarding the taxonomic position of M. leprae. The leprosy bacillus may not have any closer relationship with the representatives of the slow-growing and fast-growing mycobacteria studied. Its position in the genus Mycobacterium may therefore be separate from these other mycobacterial species. Similar results have been obtained recently by using a more conventional approach (26, 27) and are in line with earlier cytochemical investigations (9). A closer link with mycobacteria than with related genera is still suggested by studies of mycolic acids (8) as well as by earlier investigations (12-14). M. lepraemurium is the other known noncultivable mycobacterium causing disease. Studies of this bacillus have shown a close relationship with slow-growing mycobacteria, particularly with M. avium (25). Our results would suggest that the human leprosy bacillus is no more related to the mouse leprosy bacillus than to any of the slow-growing cultivable mycobacteria, including M. tuberculosis. The taxonomic validity and homogeneity of the so-called slow-growing as well as fast-growing mycobacterial groups would permit such a conclusion (26). Identification of M. leprae-specific antigens have been previously reported. Abe described a high-molecular-weight protein (NEPR) unique to the leprosy bacillus (1, 2). Stanford et al. have detected four different antigens that are found only in M. leprae (27). Since the defect in cellular immune resistance in lepromatous leprosy involves only the leprosy bacillus, it follows that the antigen(s) responsible for immunity must be M. leprae specific. Such antigens therefore deserve special attention regarding their role in humoral as well as cellular immunity in the various forms of leprosy. group
LITERATURE CITED 1. Abe, M. 1970. Studies on the antigenic specificity of Mycobacterium leprae. Demonstration of soluble antigens in leprosy modules by immunodiffusion. Int. J. Lepr. 38:113-125. 2. Abe, M., F. Minagawa, Y. Yoshino, and K. Okamura. 1972. Studies on the antigenic specificity of Mycobacterium leprae. Purification and immunological characterization of the soluble antigen in leprosy nodules. Int. J. Lepr. 40:107-117. 3. Axelsen, N. H. 1973. Quantitative immunoelectrophoretic methods as tools for a polyvalent approach to standardization in the immunochemistry of Candida albicans. Infect. Immun. 7:949-960. 4. Axelsen, N. H., J. Krall, and B. Weeke. 1973. Manual of quantitative immunoelectrophoresis. Methods and applications. Scand. J. Immunol. 2 (Suppl.):1-169. 5. Bekierkunst, A. 1968. Tubercle bacilli grown in vivo. Ann. N.Y. Acad. Sci. 154:79-87. 6. Estrada-Parra, S. 1972. Immunochemical identification of a defined antigen of Mycobacterium leprae. Infect. Immun. 5:258-259. 7. Estrada-Parra, S., S. Calder6n-Manes, M. SalazarMallen, and M.-E. Amezcua. 1966. Isolation of a group-specific polysaccharide from tissues infected with Mycobacterium leprae. Int. J. Lepr. 34:294-297. 8. Etemadi, A. H., and J. Convit. 1974. Mycolic acids from "noncultivable" mycobacteria. Infect. Immun. 10:236-239. 9. Fisher, C. A., and L. Barksdale. 1973. Cytochemical reactions of human leprosy bacilli and mycobacteria: ultrastructural implications. J. Bacteriol. 113:13891399. 10. Kirchheimer, W. F., and E. E. Storrs. 1971. Attempts to establish the armadillo (Dasypus novemcinctus Linn.) as a model for the study of leprosy. I. Report of lepromatoid leprosy in an experimentally infected armadillo. Int. J. Lepr. 39:693-702. 11. Kronvall, G., G. Bjune, J. Stanford, S. Menzel, and D. Samuel. 1975. Mycobacterial antigens in antibody responses of leprosy patients. Int. J. Lepr., in press. 12. Kwapinski, J. B. G., A. Alcasid, E. H. Kwapinski, and V. Nairn. 1972. The immunology of cytoplasmic antigens of mycobacteria. Can. J. Microbiol. 18:12011211. 13. Kwapinski, J. B. G., J. 0. de Almeida, and E. H. Kwapinski. 1972. Immunological analysis of Mycobacterium leprae using cross reactions with cytoplasmic antigens of the Actinomycetales. Can. J. Microbiol. 18:391-396. 14. Kwapinski, J. B. G., J. 0. de Almeida, and E. H. Kwapinski. 1972. Immunological determination of Mycobacterium leprae by means of cytoplasmic antigens. Bull. W.H.O. 46:509-513. 15. Laurell, C.-B. 1965. Antigen-antibody crossed electrophoresis. Anal. Biochem. 10:358-361. 16. Laurell, C.-B. 1972. Electrophoretic and electroimmunochemical analysis of proteins. Scand. J. Clin. Lab. Invest. 29:(Suppl):1-136. 17. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 18. Mudd, S. 1925. A study by new methods of the surfaces of normal and sensitized acid-fast bacteria. Proc. Soc. Exp. Biol. Med. 23:569-572. 19. Navalkar, R. G. 1971. Immunologic analysis of Mycobacterium leprae antigens by means of diffusion-in-
1138
KRONVALL, STANFORD, AND WALSH
gel methods. Int. J. Lepr. 39:105-112. 20. Navalkar, R. G. 1973. Immunologic studies on leprosy. 2. Antigenic studies of Mycobacterium leprae. Z. Tropenmed. Parasitol. 24:66-72. 21. Navalkar, R. G., P. J. Patel, R. R. Dalvi, and L. Levy. 1974. Immune response to Mycobacterium leprae: plaque-forming cells in mice. Infect. Immun. 10:13021306. 22. Ridley, D. S., and W. H. Jopling. 1962. A classification of leprosy for research purposes. Lepr. Rev. 33:119128. 23. Ridley, D. S., and W. H. Jopling. 1966. Classification of leprosy according to immunity. A five-group system. Int. J. Lepr. 34:255-273. 24. Ridley, D. S., and M. F. R. Waters. 1969. Significance of variations within the lepromatous group. Lepr. Rev. 40:143-152. 25. Stanford, J. L. 1973. An immunodiffusion analysis of
INFECT. IMMUN. 26.
27.
28. 29. 30.
Mycobacterium lepraemurium Marchoux and Sorel. J. Med. Microbiol. 6:435-439. Stanford, J. L. 1973. Immunodiffusion analysis-a rational basis for the taxonomy of mycobacteria. Ann. Soc. Belge Med. Trop. 53:321-330. Stanford, J. L., G. A. W. Rook, J. Convit, T. Godal, G. Kronvall, R. J. W. Rees, and G. P. Walsh. Preliminary taxonomic studies on the leprosy bacillus. Br. J. Exp. Pathol. 56:579-585. Storrs, E. E., G. P. Walsh, H. P. Burchfield, and C. H. Binford. 1974. Leprosy in the armadillos, new model for biomedical research. Science 183:851-852. Tuma, M., and C. Silva. 1962. Antigenic relationship between the Hansen bacillus and other mycobacteria. Int. J. Lepr. 30:71-76. Turcotte, R. 1973. Antigenic comparison between in vivo and in vitro grown tubercle bacilli. Can. J. Microbiol. 19:925-930.
